HAZARDomics addresses the metabolic perturbations in different groups of pigs (4-month old), including control groups, caused by exposure to PCBs (20 ng/kg bw/day of Arochlor 1260) and BPA (0.85-4 µg/kg bw/day of BPA) during an exposure period of 22 days. Serum samples have been investigated as the toxicological matrix and a Bligh and Dyer procedure has been applied for sample preparation. As a result, two different phases (hydrophilic and lipophilic phases) have been obtained for each sample, which were intended for metabolomics and lipidomics investigation, respectively, applying LC-HRMS as analytical tool. After data processing involving multivariate analysis [i.e. principal component analysis (PCA) and partial least square-discriminant analysis (PLS-DA)], a wide range of variables were identified as possible effect biomarkers related to PCBs exposure. Subsequent annotation of metabolites led to the identification of the tryptophan pathway as one of the metabolic pathways altered by exposure to PCBs at low concentration levels. Furthermore, the lipidomics study showed a relevant disturbance of glycerophosphocholines levels in serum caused by exposure to PCBs. On the other hand, it was also observed that lipid levels in serum were affected by exposure to low doses of BPA.
In an attempt to implement IMS technology into LC-HRMS workflows for metabolomics studies, the reproducibility of collision cross section (CCS) measurements between different IMS modes was also investigated. CCS can be considered as a novel molecular characteristic in metabolomics studies that provides additional information to retention indexes and mass spectra to support metabolite annotation. However, the complete implementation of this parameter for annotation purposes still requires fundamental research to give confidence in the reported CCS databases. Steroids (n ≈ 112 ions) were selected as metabolites of interest, since exposure to EDCs such as PCBs and BPA can alter steroid metabolism. CCS measurements were carried out with three different IMS modes, including drift tube ion mobility spectrometry (DTIMS), traveling wave ion mobility spectrometry (TWIMS) and trapped ion mobility spectrometry (TIMS). In general, deviations of less than 2% were obtained for all measurements and IMS modes compared to reference values, and which is the currently accepted threshold for CCS measurements compared to CCS databases generated with the same IMS mode.
Finally, as HAZARDOmics involved the implementation of different analytical tools in metabolomics approaches used for risk assessment purposes, gas chromatography (GC)-HRMS was applied for the detection of effect biomarkers related to PCBs exposure. In this sense, cholesterol was unequivocally identified as effect biomarker of PBCS exposure, although it was not previously detected in the metabolomics study applying the LC-HRMS workflow. This highlights the fact that metabolomics studies in risk assessment require multiplatforms that involve various analytical technologies to obtain as much information as possible on the metabolites and their concentration levels impacted by chemical exposure.
